[Tutor] create numpy array from list of strings

[Danny Yoo]

> I actually thought this would be as simple as the 'load' command to get 
> the data in from a file. But I just couldn't find the documentation 
> online to do it once i have the data already 'inside' of python as a 
> list. So, my options are:

Hello,

If you already have the values as a list of rows, where each row is a list 
of numbers, then the 'array' function may be appropriate.

     http://www.scipy.org/Cookbook/BuildingArrays

I see that you already know about array() from your second solution.



> 1) write the data block out to a temporary file and read it in using 'load'
> - which is really simple (see: http://www.scipy.org/Cookbook/InputOutput)

Alhtough it's simple, if you can avoid I/O, do so: touching disk can raise 
its own problems.  I like your second approach much better.



> 2) I've done the following, but I doubt it is the most efficient method:
>
> tempDATA=[]
> for i in data[stind:-1]:  #this is the data block from the file
>        tempDATA.append([float(j) for j in i.split()])
>
> outdata=array(tempDATA).transpose()
>
> Is there a better way?

This looks good.

I'm not sure I understand the transpose() call, although I suspect it's 
because that's the next step you want to do to process your array.




But just to follow up on what you said earlier:

> I'm not really trying to create a function.


It's actually very easy to turn what you have there into a function.  I'd 
recommend doing so.  Here's what it looks like.

#######################################################
def convert(data):
     tempDATA = []
     for i in data:
         tempDATA.append([float(j) for j in i.split()])
     return array(tempDATA)
#######################################################

To get back the same behavior as your code above, you can call this 
function as:

     outdata = convert(data[stind:-1]).transpose()

Functions give you a way to bundle up a collection of related operations. 
More importantly, that 'tempDATA' variable doesn't stick around hanging 
out in space: it only lives in the convert() function.